Scientific papers

Peer-reviewed and published.

Optain’s technology is backed by first-class research published in leading academic journals and peer-reviewed by world-renowned experts in AI, ophthalmology, and retinal imaging.

nAMD
Development and validation of a deep-learning algorithm for the detection of neovascular age-related macular degeneration (AMD) from color fundus photographs

Diabetic retinopathy
An Automated Grading System for Detection of Vision-Threatening Referable Diabetic Retinopathy on the Basis of Color Fundus Photographs

Glaucoma
Efficacy of a Deep Learning System for Detecting Glaucomatous Optic Neuropathy Based on Color Fundus Photographs

Cardiovascular disease
A Deep Learning System for Fully Automated Retinal Vessel Measurement in High Throughput Image Analysis

Research and Publications

Opthalmic diseases
Using deep leaning models to detect ophthalmic diseases: A comparative study

Artificial intelligence
Ophthalmology and the emergence of artificial intelligence

Diabetic retinopathy and glaucoma
Artificial Intelligence in Ophthalmology: Accuracy, Challenges, and Clinical Application

Artificial intelligence
Visualizing Deep Learning Models for the Detection of Referable Diabetic Retinopathy and Glaucoma

Artificial intelligence
Development and Validation of a Deep Learning System for Diabetic Retinopathy and Related Eye Diseases Using Retinal Images From Multiethnic Populations With Diabetes

Diabetic retinopathy
Artificial Intelligence Screening for Diabetic Retinopathy: the Real-World Emerging Application

FAQs

FAQ: Results and Diseases 

Eyetelligence Assure utilises deep learning to scan retinal images for features of eye disease. It presents these findings as low, medium, or high risk in the case of glaucoma, and as low or high risk for referrable DR and referrable AMD. Its purpose is to give you more information that you can consider and verify, in the context of other information, such as patient case history, and to determine if additional testing may be warranted. 

The software is most useful in the assessment of new patients to a practice or where a patient’s presentation has changed and is especially helpful in clinics with high patient traffic where mistakes can occur, and conditions can be missed. Eyetelligence is designed to draw a practitioner’s attention in cases where features of eye disease have been detected. In cases where a diagnosis has already been made, and the management of the patient has been determined, Eyetelligence cannot add value. 

The software grades cases of diabetic retinopathy according to the International Council of Ophthalmology Guidelines for Diabetic Eye Care (2017) and the NHS Diabetic Eye Screening Grading Definitions. Consistent with the RANZCO Screening and Referral Pathway for Diabetic Retinopathy, this grading system determines whether a patient should be monitored or referred to an ophthalmologist. 

A grey or ‘ungradable’ result means that the image was not suitable for screening for a particular condition. You may wish to adjust the patient’s position and retake the image. 

The role of Eyetelligence Assure is to be a clinical decision support tool. It is not intended to report a categorical diagnosis of eye disease. The software provides additional information, regarding the interpretation of ophthalmic images, to support optometrists in making decisions pertaining to diagnosis and management. 

Glaucoma detection is challenging, even for a specialist. While you never want to miss a true glaucoma patient, you also don’t want to perform visual field tests (VFT) unnecessarily. Eyetelligence Assure allows you to triage patients and identify those who need a VTF. 

The approved use for this application is for use in people over the age of 40. Results for patients under 40 years of age should be interpreted with caution due to RNFL reflection and artefacts. 

It is important to note that Eyetelligence Assure does not determine features of early, intermediate AMD, or geographic atrophy AMD; only features consistent with late wet (referrable) AMD. 

FAQ: AI and Research

AI will not replace the clinician; it is not designed to do so. Its function is to augment and support your clinical decision-making. The Eyetelligence Assure software interprets retinal images to give you more information. To accurately diagnose glaucoma, you need more supporting evidence, and we are interpreting an image modality that has never had diagnostic support before. It’s another piece in the puzzle, to help you make decisions more swifty and know what you need to do next. 

Yes, each of our AI disease models in the Eyetelligence Assure software has research papers which have been peer-reviewed and published, you can find them here in our Research section. 

AI (artificial intelligence) or deep learning, is the broad definition given to machine learning technology that has the ability to be trained to recognise various data sets and patterns, in order to build and learn knowledge to make decisions. Our software is an AI-based deep learning model that was trained using a data set of millions of images across the three major blinding eye diseases, assessed and labelled by a panel of ophthalmologists without clinical context, case history or other investigations to assist them. A label was determined when a consensus of three panellists was reached. In the software, when it’s shown a new fundus image, it assigns a label based on the culmination of this learning. 

Sensitivity is the proportion of cases that a test detects a condition when it is actually present. For example, suppose there is a test that detects blue eyes in the population. 100 people are checked, 10 of which have blue eyes. If the test detects 9 out of the 10 with blue eyes, it has a sensitivity of 90%. Specificity is the proportion of cases that are detected by a test as having a condition that actually have the condition. Using the same example, if the test was to detect 10 people with blue eyes from the group of 100 people but only 9 of those detected actually had blue eyes, its specificity is 90%. 

FAQ: Installation and set-up

If your Eyetelligence Assure software is installed on a computer connected to the internet, simply login to the software using your email and password. If your computer is offline, you will be asked to scan a QR code with your mobile phone. On your phone, you’ll be asked to input your email, password, and computer ID, which will be provided to you automatically. 

Eyetelligence Assure has a feature called Web Server. Once this has been activated under ‘Settings’ it will provide you a URL web address. Simply copy and paste or write this URL web address into another computer’s web address bar (examples web browsers are Chrome and Internet Explorer). The Eyetelligence Assure ‘Results’ page will be listed along with the ‘fundus grading report’ under the ‘View’ button. 

Eyetelligence Assure may only require an export function from image viewing software. For more information, please visit this Image Set-up page. 

  • Right click on the Eyetelligence file in your Downloads folder. 
  • Click ‘Extract All’. 
  • Double click on ‘EyetelligenceAssure’. 
  • Run as Administrator. 

If you have any questions, please contact it-support@eyetelligence.com.au 

Eyetelligence Assure is compatible with most retinal and OCT cameras. 

Eyetelligence Assure is compatible with computers operating only Windows systems. It will require a minimum of 4GB RAM. 

FAQ: Plans

No, we offer both monthly and annual plans. Annual plans are billed upfront (monthly price is shown for comparison). By selecting an annual plan, you will receive 20% off your subscription. 

You can cancel your plan at any time through your account dashboard. Once cancelled you will not be charged at your next billing period. You will be able to access Eyetelligence Assure for the remainder of your billing period. No refunds will be issued for the remaining time. 

Yes, you can upgrade, downgrade, or cancel your plan at any time through your account dashboard. 

You can select between 1-5 cameras during the trial signup process and can add more licenses to your trial through your account dashboard. Once your trial is over you will be able to select the plan that best suits you. 

FAQ: Technical

There are several reasons this may occur. Firstly, the folder set up where the retinal images are exported may be too large. As a consequence, we recommend deleting image files in this folder every 2 weeks. This folder can be found in C Drive  Eyetelligence   Eyetelligence Images. 

Secondly, please check your computer hardware. If you are using an operating system that is older than Windows 7, then this may be slowing down the software and potentially causing it to freeze at times. 

Lastly, you can check your RAM. If you have less than 8GB of RAM, then this may be this issue. 

The Eyetelligence Assure software is intended to be used prospectively and not retrospectively. If there are confirmed clinical cases of eye disease with the inclusion of additional medical parameters, then the Assure software will not provide any further use in your assessment. The software’s purpose is designed to assist the optometrist where patient cases are difficult to detect upon initial consultations. 

You may download the report to your computer and attach it to an email or use the import function attached to an Optimate/Sunix patient record. 

Eyetelligence Assure is compatible with most retinal and OCT cameras. 

Eyetelligence Assure is compatible with computers operating only Windows systems. It will require a minimum of 4GB RAM. 

Some clinicians choose to export retinal images into practice management software such as Optimate or Sunix. Once an image has been exported into such programs, Optimate and Sunix will clear the images from the folder they have been located. If this folder is the same as the Image Folder placed under ‘Settings’, then the Eyetelligence software can no longer display the images and results. It is important to note that the Eyetelligence Assure software does not store any data or images, it simply reads image files stored on the computer folders. 

This may be a result of either one or both of two issues; firstly, Eyetelligence Assure must be launched prior to capturing a retinal image in order for the software to detect the image. Secondly, it is important to note that the upload of new images to the software must be prepared ready for use. Any existing images from past patients must be cleared from the software interface. Please click the ‘Reset’ button to prepare a new analysis for the next patient. 

Eyetelligence Assure has a feature called Web Server. Once this has been activated under ‘Settings’ it will provide you a URL web address. Simply copy and paste or write this URL web address into another computer’s web address bar (examples web browsers are Chrome and Internet Explorer). The Eyetelligence Assure ‘Results’ page will be listed along with the ‘fundus grading report’ under the ‘View’ Button. 

Eyetelligence Assure may only require an export function from image viewing software. For more information, please visit this Image Set-up page. 

FAQ: Confidentiality and security

Eyetelligence Assure is a secure online solution employing stringent access controls to ensure patient data is never compromised.  

Fundus images used in the AI training are de-identified and made untraceable in line with the Safe Harbour Data Method. All Personally identifiable information (PII) and protected health information (PHI) are removed to hide the patient’s identity before specialist graders systematically assess the images for training. 

Regulatory approval Statements

Class 1 Medical Devices meeting international standards.
Optain AI-enabled software and fundus cameras are classified as a Class 1 medical device, registered with the Australian TGA, and carry European CE Marking, conforming to the performance and safety standards in over 30 countries across the EU and Australasia.

 

Eyetelligence Assure

AI Software for Eye Disease Screening 

  • Aust: TGA Class 1 Medical Device 
  • EU: CE mark Class 1 Medical Device 
  • UK: MHRA Class 1 Medical Device 
  • NZ: Medsafe Class 1 Medical Device 

Eyetelligence Assure+

AI Software for Eye Disease Screening + Cardiovascular Assessment 

  • Aust: TGA Class 1 Medical Device 
  • EU: CE mark Class 1 Medical Device 
  • UK: MHRA Class 1 Medical Device 
  • NZ: Medsafe Class 1 Medical Device

Resolve Camera

Portable Automatic Fundus Camera 

  • US: FDA Class 1 Medical Device 
  • Aust: TGA Class 1 Medical Device 
  • EU: CE mark Class 1 Medical Device 
  • UK: MHRA Class 1 Medical Device 
  • NZ: Medsafe Class 1 Medical Device